Brake Pad Wear Estimation from Braking Event Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current brake pad wear monitoring systems face challenges in accurately determining wear, especially when using worn brake pads of different types, due to high measurement uncertainties and the need for recalibration, and are costly due to the use of additional sensors and hardware.
Innovation Solution
A method using time series data from vehicle sensors to identify and classify braking events, employing machine learning models like logistic regression to estimate brake pad wear based on features derived from statistical operators, eliminating the need for direct thickness measurement sensors and allowing for the use of brake pads from various manufacturers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hardware sensors are used for direct measurement of brake pad thickness, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces mechanical/direct sensing systems with an indirect measurement approach using existing vehicle sensors (accelerometers, gyroscopes, pressure sensors) to detect braking events and calculate pad wear through signal analysis, eliminating the need for dedicated thickness measurement sensors
Solution Approach 2:
The patent introduces an intermediary measurement approach where brake pad wear is not measured directly but inferred through intermediate parameters such as braking event characteristics, pedal force, and temporal signal analysis of existing sensor data
2Measurement precision
If brake pad wear is monitored using integral approach with BTM, then wear estimation is achieved, but measurement uncertainty increases continuously
Solution Approach 1:
The patent segments the continuous wear estimation problem into discrete braking event analyses, where each braking event is independently detected and classified, preventing error accumulation that occurs in continuous integral approaches
Solution Approach 2:
The patent focuses analysis on specific partial aspects of braking events (temporal characteristics, force profiles, acceleration patterns) rather than attempting to model all wear-influencing factors, achieving reliable wear detection without the complexity and uncertainty of comprehensive modeling
3Measurement precision
If multi-stage BPWS are used for recalibration and safety monitoring, then wear classification is improved, but maintenance costs increase due to sensor replacement
Solution Approach 1:
The system uses existing vehicle sensors and onboard processing capabilities to perform wear monitoring and classification functions, eliminating the need for dedicated wearable sensors that would require replacement with pad changes
Solution Approach 2:
The patent makes existing vehicle sensors (accelerometers, pressure sensors, gyroscopes) perform the additional function of brake pad wear detection, eliminating the need for separate dedicated sensors and reducing overall system complexity
4Measurement precision
If BTM with hardware sensors for brake disk temperature is used, then temperature measurement accuracy is improved, but product costs increase greatly
Solution Approach 1:
The patent uses existing vehicle sensors (accelerometers, pressure sensors, wheel speed sensors) to infer temperature-related wear conditions without requiring dedicated temperature sensors, achieving wear monitoring functionality through multi-purpose use of available sensors
Data Source
AI summary
A method for determining a state of wear of a brake pad of a vehicle. The method includes: receiving time series data, the time series data including a time series of brake system-related data of the vehicle; identifying at least one braking event in the time series data, each braking event identified in the time series data corresponding to a temporal data window of braking event data of the time series data, the data window correlating with a real braking event of the vehicle; determining features from the braking event data by using predetermined operators for each identified braking event; classifying the at least one braking event by using the features determined for this purpose, the classification being associated with a state of wear of the brake pad of the vehicle.